The role of LBD proteins in floral organ abscission

(1)

The role of LBD proteins in floral organ abscission

Ingerid Ørjansen Kirkeleite

Thesis for the Degree Master of Science 60 study points

UNIVERSITY OF OSLO

Faculty of Mathematics and Natural Science

Department of Biosciences

(2)

(3)

I The work presented in this thesis was carried out at the Department of Biosciences, Faculty of Mathematics and Natural Sciences University of Oslo, in the period between January 2013 and August 2014, Supervision has been provided by Professor Reidunn B. Aalen (formal supervisor), Research fellow Melinka A. Butenko and Post doc. Chun-Lin Shi.

First I would like to thank Professor Reidunn B. Aalen for the opportunity to do my master thesis in her research group and Research fellow Melinka A. Butenko for including me in her research project. I am grateful for all the skilled guidance and supervision in the process of working out my master thesis.

Many thanks to Post doc. Chun-Lin Shi for all the help in the lab and during the writing of my thesis.

Thanks to Ph.D. student Mari Wildhagen for reading my master thesis and for always being so cheerful. Thanks also to Post doc. Ullrich Herrmann for gladly lending a hand and for all the nice conversations.

I wish to thank Solveig Hauge Engebretsen and Roy Fallet for always lending a smile and a helping hand in the lab and in the phytotron.

A special thanks to Line and the other master students with whom I have had the pleasure of sharing an “office”. They have made this journey so much easier.

Finally, I want to thank all my friends and family for all their love and support, and last, but not least, special thanks to Lars-Inge, my boyfriend, for all the smiles, jokes and encouraging words I so desperately needed the last couple of weeks.

Oslo, August 2014 Ingerid Kirkeleite

(4)

(5)

II Abscission is a developmental process where plant organs are actively shed from the plant body either due to infection or simply because the organ no longer serves a purpose for the plant. In Arabidopsis thaliana (Arabidopsis) floral organs are shed after pollination in a cell separation event that has been shown to be controlled by the INFLORECENCE DEFICIENT IN ARABIDOPSIS (IDA) peptide through the two receptor-like kinases, HAESA and HAESA-LIKE2. IDA activates a MITOGEN-ACTIVATED PROTEIN KINASE (MAPK) cascade inducing cell separation in the abscission zone (AZ). The MAPK cascade, again, is believed to regulate three Arabidopsis KNOX homeodomain transcription factors (TF), KNAT1 which inhibits cell wall loosening and controls KNAT2 and KNAT6 which induce cell separation. During leaf development KNOX genes have been implicated to be directly repressed by a complex consisting of ASYMMETRIC LEAVES 1 a member of the HLH family and ASYMMETRIC LEAVES2, a member of the LATERAL ORGAN BOUNDARIES DOMAIN (LBD) family. Another LBD protein, JAGGED LATERAL ORGAN (JLO), is also involved in KNOX regulation when coordinating organ development in shoot and floral meristems. JLO and AS2 can together with AS1 form trimeric complexes to suppress KNAT1 expression during lateral organ development. We recently showed that two other members of the LBD family, LBD37 and LBD39, are up-regulated in ida and hae hsl2 mutants, indicating that members of the LBDs are being down-regulated in the IDA signalling pathway. We propose that IDA signalling negatively regulates the LBDs in abscission zone cells, preventing the LBDs from down-regulating KNAT2 and KNAT6, this in turn allowing these TFs to induce floral organ abscission. Here, we will explore the possibility of this proposed model by investigating single and higher order mutants of four closely related LBD genes, including LBD37 and LBD39, their expression pattern during floral organ abscission, and by determining their genetic interactions with KNAT2, KNAT6, IDA and HAE HSL2.

(6)

II

(7)

1 Introduction

1.1 Arabidopsis as a model organism

Arabidopsis is a small robust flowering plant from the mustard family (Meinke et al., 1998). Due to its physical traits, short generation time (about six weeks), self- pollination and its ability to produce a large number of seeds Arabidopsis has been a first choice of study of many plant biologists. In addition, Arabidopsis is easily transformed by the Agrobacterium tumefaciens, which can transfer its T-DNA into the plant cell where the T-DNA integrates into the plant genome (Clough et al., 1998).

In 2000 Arabidopsis was the first plant ever to get its genome fully sequenced.

The genome consists of 125Mega bases constituting one of the smallest genomes in the plant kingdom (Arabidopsis Genome Initiative, 2000).

The genome is organized into five chromosomes containing around 25,500 genes (Arabidopsis Genome Initiative, 2000).

Arabidopsis has been shown to be a suitable model organism for investigating an extensive number of essential developmental processes in plants. Abscission, the programmed loss of plant organs, is important during plant development and Arabidopsis has proven to be a suitable system for studying this process as the characteristics for floral organ abscission are similar to that of other abscission processes found in other plant species like beans and tomatoes (Bleecker and Patterson, 1997; Patterson, 2001). In addition, the presence of progressive older flowers and siliques from unfertilized buds to mature siliques along the length of a single inflorescence makes Arabidopsis useful for following the different stages of floral abscission and the changes that the cells undergoing separation go through.

(12)

2

1.2 Cell separation in plants

A number of cell separation events take place at different developmental stages during the life cycle of a plant and are caused by the breakdown of the pectin rich matrix that connects adjacent cells preceded or followed by cell wall remodeling (Roberts et al., 2002; Jarvis et al., 2003). Cell separation is a process important for the plants reproductive success, shattering of infected organs or organs that have served their purpose for the plant as well as for architectural alterations of the plant body. The different sites at which cell separation can occur are displayed in figure 1.1 and include shattering of pods and shedding of germinated seeds as well as the seed germination itself, pollen release, shedding of organs and different tissues, and emergence of lateral roots (Roberts and Roberts, 2000; Roberts et al., 2002;

Swarup et al., 2008). Cell separation has also been observed at the root tip where the peripheral cells are sloughed off as the root makes its way through the soil (del Campillo et al., 2004).

The physiological aspects taking place during the different cell separation processes share common cellular and cell wall changes and the signaling processes initiating the different separation events are believed to be conserved. Still, there are likely different environmental and biological regulatory factors that initiate the different cell separation processes so that cell separation only occurs at specific locations at suitable times. From an agricultural perspective delayed cell separation has always been an advantageous trait for avoiding preharvest shedding of fruits and seeds and to prolong the time pollination can take place to increase crop yields and quality (Pickersgill, 2007).

(13)

3 Figure 1.1: The different sites where cell separation processes may occur in plants.

(Roberts et al., 2002).

(14)

4

1.2.1 Abscission

Abscission is the term used for processes leading to plants shedding their organs. Organ abscission is important for the plant in the removal of infected, dead or non-functional organs, but also healthy organs no longer serving a function as well as shedding of fruit and seeds (Sexton et al., 1982; Bleecker and Patterson, 1997). The position at which organ shedding takes place is called the abscission zone (AZ) (Sexton et al., 1982; González Carranza et al., 1998; Taylor et al., 2001). The AZ is located between the shedding organ and the plant body.

It is characterized as several cell layers consisting of small cytoplasmically dense cells that fail to enlarge like the surrounding cells until developmental and environmental conditions trigger the initiation of abscission (Bleecker and Patterson, 1997; Roberts et al., 2002). Once triggered, there is an up-regulation of genes encoding hydrolyzing enzymes like expansins (EXP) and xyloglucan endotransglucosylase/hydrolase (XTH) which lead to a restructuring of AZ cell walls allowing for an expansion of the cells (Cosgrove, 1998; Cai and Lashbrook, 2008; Ogawa et al., 2009). Subsequently pectin-hydrolysing enzymes like polygalacturonase (PG) cause dissolution of the middle lamella as pectin gets demethylated causing a fracture plane in the separating layer (Sexton et al., 1982; González-Carranza et al., 2007; Cai and Lashbrook, 2008; Ogawa et al., 2009).To avoid pathogen invasion and loss of water and nutrients a corky layer forms across the stem as the organ detaches (Patterson, 2001).

Changes in the levels of hormones like ethylene and auxin are important in regulation of the timing of abscission as ethylene activates abscission while auxin delays abscission by making the AZ cells less responsive to ethylene (Osborne et al., 1989; Taylor et al., 2001). However, indoleacetic acid (IAA), a member of the auxin class of plant hormones, is not only important for timing but also for the rate at which the cell wall is degraded and the need for a functional IAA signaling pathway for organ abscission to take place have recently been demonstrated (Basu et al., 2013). In lateral root emergence (LRE) auxin influx has been shown to be the starting point of the process of separating cell layers overlaying the incipient lateral root further expanding the role of auxin in cell separation processes (Kumpf et al., 2013).

(15)

5

1.2.2 Floral organ abscission

Floral organ abscission is the process where floral organs like petals, stamen and sepals are abscised from the floral AZ, which is believed to derive from the floral meristem (Gawadi, 1950; Patterson, 2001; van Nocker and van, 2009). In Arabidopsis the floral organs are shed after pollination at a time where the floral organs have served their purpose for the plant (Patterson, 2001). The knowledge of the different genes regulating the process of floral abscission is steadily increasing, and has challenged the role of classical plant hormones, like ethylene and auxin, as the most important regulators of plant development (Butenko et al., 2009; Stahl et al., 2010).

In Arabidopsis BLADE-ON-PETIOLE1 (BOP1) and BOP2, two NPR1-like signaling proteins with conserved BTB/POZ domain and ankyrin repeats (Hepworth et al., 2005;

McKim et al., 2008) have been found to play a role in promoting differentiation of floral organ AZ cells (figure 1.2) as loss of BOP1 and BOP2 function prevents proper structural formation of AZ cells and bop1 bop2 show no sign of weakening of the petals (McKim et al., 2008).

The INFLORESCENCE DEFICIENT IN ABSCISSION (IDA) peptide has been found to be involved in the final step of abscission (Figure 1.2) where the actual cell separation by dissociation of cell walls occurs as the ida mutant show maturation of the AZ, but organ shedding does not take place (Butenko et al., 2003). IDA is expressed shortly after sensitization of the AZ and seems to be involved in cell wall loosening prior to organ detachment, and display precocious floral organ abscission and an enlargement of AZ cells at the point of organ detachment when overexpressed (Butenko et al., 2003; Butenko et al., 2006; Stenvik et al., 2006). Earlier and epistatic expression of IDA could not promote abscission in bop1 bop2 (McKim et al., 2008). Thus, the presence of atomically differentiated AZ cells most likely provided by BOP1 and BOP2 are essential for the IDA peptide to promote floral organ abscission (McKim et al., 2008; Shi et al., 2011).

Another gene called NEVERSHED (NEV), a ADP-ribosylation factor GTPase-activating protein located in the trans-Golgi network, is thought to act at the same stages as IDA regulating abscission (Figure 1.2) (Liljegren et al., 2009). Loss of NEV displays deficient abscission of floral organs and causes like overexpression of IDA ectopic expansion of the AZ cells (Liljegren et al., 2009; Liu et al., 2013). Transmission electron micrographs revealed

(16)

6

altered Golgi structure and location of the Golgi network in the sepal AZs of nev mutants and genes suggested to be involved in cell separation were found to be downregulated in nev mutants (Liu et al., 2013). Thus NEV likely regulates membrane trafficking sustaining a proper organization of the Golgi apparatus and correct localization of the Golgi network important for the progression of abscission and cell separation (Liljegren et al., 2009; Stefano et al., 2010). Taken together NEV seems to act as an inhibitor of cell expansion while IDA promotes it, whereas both NEV and IDA are needed for the promotion and proper execution of cell separation(Butenko et al., 2003; Butenko et al., 2006; Stenvik et al., 2006; Liljegren et al., 2009; Liu et al., 2013).

HAESA (HAE) and HAESA-LIKE 2 (HSL2), two leucine-rich repeat-receptor-like kinase (LRR-LRK), are involved in the promotion of abscission as receptors for the IDA peptide transducing its signal to cytoplasmic effectors (Cho et al., 2008; McKim et al., 2008; Stenvik et al., 2008; Butenko et al., 2009; Butenko et al., 2014). EVRSHED (EVR), CAST AWAY (CST) and SOMATIC EMBRYOGENESIS RECEPTOR-LIKE KINASE1 (SERK1) on the other hand are the three receptor-like kinases implied to work as spatial inhibitors of abscission by preventing abscission activation (Leslie et al., 2010; Lewis et al., 2010; Burr et al., 2011). Mutations in EVR, CST and SERK1 have the ability to restore the trans-Golgi network and organ abscission in nev (Leslie et al., 2010; Lewis et al., 2010; Burr et al., 2011).

In fact mutations of CST, EVR and SERK1 in nev lead to precocious abscission and enlargement of AZ layer supporting their role as negative regulators of abscission (Leslie et al., 2010; Lewis et al., 2010; Burr et al., 2011). However, evr, serk1 and cst alone does not display any visible abscission defects and were unable to rescue ida and haehsl2 abscission defects suggesting they act in a different pathway than IDA (Leslie et al., 2010; Lewis et al., 2010; Burr et al., 2011). Even so, it has been suggested that EVR, CST and SERK1 are able to block the IDA signal by forming a complex with HAE HSL2 prior to ligand binding triggering internalization and recycling of the receptors via endocytosis regulated by the NEV molecule (Liljegren et al., 2009; Burr et al., 2011; Liu et al., 2013).

BREVIPEDICELLUS (BP)/ KNOTTED-LIKE FROM ARABIDOPSIS THALIANA1 (KNAT1) a member of the KNOTTED-LIKE HOMEODOMAIN (KNOX) gene family, are proposed to play an important role in proper timing and regulation of the enlargement of morphologically distinct AZ cells in floral organ abscission (Figure 1.2) (Wang et al., 2006; Shi et al., 2011).

Loss of KNAT1 activity, as with gain of IDA function leads to enlargement of AZ cells and

(17)

7 early cell wall loosening at the positions where shedding of the floral organs take place (Shi et al., 2011). Two other KNAT genes (KNAT2 and KNAT6) are believed to be promoters of abscission by direct activation of Cell Wall Remodeling (CWR) genes and are in turn suggested to be regulated by BP/KNAT1 (Ragni et al., 2008; Shi et al., 2011). KNAT1 may also restrict cell expansion and proliferation through activation of EVR, as EVR is downregulated in knat1 mutants (Shi et al., 2011), though, likely in an IDA independent manner suggesting the involvement of other genes in the interconnected pathways (Leslie et al., 2010).

Figure 1.2 Model of floral organ abscission in Arabidopsis. The events and the genes acting in floral organ abscission from AZ formation to cell separation are demonstrated here.

On the left the genes involved in initiation of AZ formation and differentiation, activation of cell wall loosening, dissolution of middle lamella are listed. Genes involved in the latter remain active or actively repressed throughout the cell separating process. (Gray boxes mark inhibitors of abscission.) On the right the physiological aspects is briefly described also including some of the central enzymes enacting in cell wall loosening and dissolution of the middle lamella. Modified from Patterson (2001), Liljegren et al., (2012) and Aalen et al., (2013).

(18)

8

1.3 IDA signaling pathway

Many of the different genes found to act in floral organ abscission also act in the same pathway. One of the most well studied signaling pathways leading to loss of floral organs as well as cell separation during LRE is the IDA signaling pathway where IDA and its HAE HSL2 receptors constitute the starting point (figure 1.3) (Cho et al., 2008; Stenvik et al., 2008;

Butenko et al., 2009). IDA was found during a screening for delayed abscission as a recessive gene encoding a peptide regulating floral organ abscission in an ethylene-independent manner (Butenko et al., 2003). Lack of functional IDA peptide causes all the floral organs to remain attached to the abscission zones of the siliques (Butenko et al., 2003). HAE and HSL2 are expressed at the AZ of floral organs in a pattern similar to that of the IDA peptide and show the same phenotype when mutated as the ida mutant (Jinn et al., 2000; Butenko et al., 2003).

IDA is highly likely a small post-translationally modified peptide secreted from the AZ cells with an ability to bind HSL2 and HAE although cells of the responding tissue might need higher concentrations of HAE due to its low affinity (Butenko et al., 2003; Stenvik et al., 2006; Stenvik et al., 2008; Matsubayashi, 2011; Butenko et al., 2014). Proper downsizing and modifications of the 77 aas long polypeptide encoded by the IDA gene is required for gaining a fully active peptide (Matsubayashi, 2011). PIP, a 12 aas long part of the 20 aas long conserved proline-rich C-terminal domain of IDA termed the Extended PIP (EPIP) motif and even more so hydroxylated PIP on at least the 7^th pro have been shown to have high affinity for the HSL2 receptor in a bioassay using oxidative burst (Butenko et al., 2014). PIP and PIPPo could not as efficiently activate HAE and further modifications of IDA or even the presence of a coreceptor may be necessary for binding to the HAE receptor (Butenko et al., 2014).

Genes encoding INFLORESCENCE DEFICIENT IN ABSCISSION-LIKE (IDL) 1-5 peptides have been found during bioinformatic screenings all sharing a conserved EPIP motif and are therefore also likely to act as signaling peptides binding RLKs (Stenvik et al., 2008).

IDL genes are like IDA expressed in AZ, however, IDL expression peaks after organ shedding in contrast to IDA suggesting separate functions from IDA (Stenvik et al., 2008).

HAE HSL2 are thought to initiate transduction of the IDA signal via a Mitogen-Activated Protein Kinase (MAPK) cascade in the cytoplasm (figure 1.3) that includes Mitogen- Activated Protein Kinase4 (MKK4) and MKK5 and their two known targets Mitogen-

(19)

9 Activated Protein Kinase 3(MPK3) and MPK6 (Tena et al., 2001; Cho et al., 2008; Kim et al., 2011). MKK4 and MKK5 when constitutively expressed could rescue the phenotype of ida and hae hsl2 and MPK6 transgene together with mpk3 mutant showed deficiency in abscission giving them a role as positive regulators of abscission in the IDA-HAE HSL2 pathway (Cho et al., 2008). The activation of MAPK cascade is believed to directly or indirectly down regulate KNAT1 found to function downstream of IDA and HAE HSL2 (Shi et al., 2011). Genetic interaction studies suggest a role of KNAT1 as a negative regulator of abscission acting downstream of IDA peptide and HAE HSL2 receptors negatively regulating KNAT2 and KNAT6 (figure 1.3) (Shi et al., 2011). Both knat2knat6 and bp-3ida-2knat2knat6 mutants show delayed abscission and the absence of KNAT6 and KNAT2 can rescue the knat1 phenotype of downward-pointing siliques supporting a role for KNAT2 and KNAT6 downstream of KNAT1 (Ragni et al., 2008; Shi et al., 2011). Furthermore, the expression level of KNAT2 and KNAT6 are elevated in knat1 mutants, while in ida and haehsl2 mutants the expression levels are almost abundant compared to wild type (Shi et al., 2011). This further supports the hypothesis of KNAT2 and KNAT6 acting as positive regulators of abscission by induction of Cell Wall Remodeling (CWR) genes as a downstream effect of the action of IDA and HAE HSL2 (Shi et al., 2011).

More recent studies have shown that IDA-HAEHSL2 signaling also is important in the cell separating processes of LRE (Kumpf et al., 2013). Kumpf et. al. (2013) demonstrated that auxin influx could induce IDA in overlaying endodermal (EN), cortical (CO) and epidermal (EP) layers which then by signaling through HAE HSL2 caused activation of CWR genes encoding cell wall remodeling enzymes causing dissolution of that middle lamella between cells of the EN, CO and EP layers enabling the emergence of the LR (Kumpf et al., 2013).

(20)

10

Figure 1.3: The IDA signaling pathway. The IDA signaling pathway so far involves signaling by IDA through HAE HSL2 causing activation of a MAP cascade that leads to the repression of KNAT1 leaving KNAT2 and KNAT6 active and able to activate transcription of CWR genes. The stippled line represents the uncertainties involving the regulation of KNAT2 and KNAT6 by KNAT1. The figure is modified from Shi et al., (2011).

1.4 The LBD family of transcription factors

In addition to their involvement in floral organ abscission, BP/KNAT1, KNAT2 and KNAT6 have previously been demonstrated to be required for meristem maintenance and organ patterning (Bürglin, 1997; Müller et al., 2001). Recent studies have found several members of the LATERAL ORGAN BOUNDARIES Domain (LBD) Gene Family to be involved in the regulation of the same KNOX genes during plant development (Semiarti et al., 2001; Borghi et al., 2007; Guo et al., 2008; Rast and Simon, 2012). LBDs, which often work as heterodimers, have been shown to act as transcription factors in various developmental processes controlling plant architecture and growth working with other LBDs when regulating their targets (Semiarti et al., 2001; Shuai et al., 2002; Xu et al., 2003; Borghi et al., 2007; Rast and Simon, 2012). In the last decade several LBDs have been characterized further increasing the knowledge about the LBD family.

(21)

11 The LBD family of Arabidopsis consists of 43 members, all of which contain a 100 amino acid long Zinc finger-like domain with a conserved 4 Cys motif (CX₂CX₆CX₃C) and a conserved Pro residue constituting the LATERAL ORGAN BOUNDARIES domain (LOB- domain) (figure 1.4) (Iwakawa et al., 2002; Shuai et al., 2002). The LBD family is a plant specific family of transcription factors divided into Class I and Class II where 37 of the total 43 members constitute Class I where the similarities between the members are generally higher that between the six members of Class II (Shuai et al., 2002).

33 of the 37 members in the Class I of LBD genes are predicted to form a coiled-coil structure at the end of the LOB domain where four Leucines with the spacing LX6LX3LX6L resembling a Leucine Zipper are found (Iwakawa et al., 2002; Shuai et al., 2002). Leucine Zippers in general are involved in protein-protein interactions suggesting that the coiled coil of the LBDs also mediate association with other proteins (Ellenberger et al., 1992). None of the Class II members, however, are predicted to have this coiled-coil suggesting a different function. An expression study of all the predicted LBD gene family members (Shuai et al., 2002), shows expression of LBD genes in a variety of tissues suggesting the involvement of LBDs in many different plant specific processes as LBDs are only found in plants (Shuai et al., 2002;

Kawade et al., 2013). Further supporting the diverse roles of LBDs in plant specific development are their involvement in embryogenesis, organ formation and organ boundary definition (Shuai et al., 2002; Xu et al., 2003; Borghi et al., 2007; Okushima et al., 2007; Rast and Simon, 2012).

The LOB domain has a DNA binding activity that binds cis-element 5’-GCGGCG-3’ giving LBD proteins the ability to form dimers with other LBDs as well as basic Helix-Loop-Helix (bHLH) family members, supporting the involvement of LBDs in protein-protein interactions (Husbands et al., 2007; Guo et al., 2008; Rast and Simon, 2012). This activity seems conserved through the entire LBD family of Arabidopsis as even LBD41 from Class II has this activity (Husbands et al., 2007).

Phylogenetic analysis showed high similarity in the LOB domain of various LBDs and RT- PCR revealed overlapping expression indicating redundancy amongst the LBDs (Shuai et al., 2002). However, one LOB domain cannot substitute a LOB domain of another closely related LBD as a LOB domain swap revealed that varying amino acid within the domain are responsible for the function of the different LBDs (Mangeon et al., 2012).

(22)

12

1.4.1 LBDs negatively regulate Class I KNOTTED-Like Homeobox Proteins

Plant meristems are crucial for plant growth and organ development at early and later developmental stages. KNOX proteins are found to be involved in meristem maintenance actively preserving a pool of undifferentiated cells residing within the meristems from which organ primordia derive (Barton and Poethig, 1993; Long et al., 1996; Byrne et al., 2002).

Upon organ formation and differentiation, repression of KNOX genes in cells determined for differentiation is important for the development of organs (Byrne et al., 2000; Ori et al., 2000;

Semiarti et al., 2001; Xu et al., 2003). ASYMMETRIC LEAVES 2 (AS2) and JAGGED LATERAL ORGANS (JLO) are two LBDs found to act as negative regulators of KNOX genes important for the establishment of boundaries between organs (figure 1.5) (Semiarti et al., 2001; Byrne et al., 2002; Shuai et al., 2002; Xu et al., 2003; Borghi et al., 2007; Guo et al., 2008; Rast and Simon, 2012). AS2 is expressed in leaf primordia where it repress KNAT1 during leaf development giving developing leaves their adaxial identity (Lin et al., 2003;

Mele et al., 2003), by acting in a complex with ASYMMETRIC LEAVES 1, a MYB domain containing transcription factor ( and a member of the HLH family) (Barton and Poethig, 1993;

Figure 1.4: LOB domain of class II LBDs in comparison to the LOB domain of LOB from Class I. The gray area marks the LOB domain of the class II LBD members (LBD37 - LBD42) and the LOB of class I, the C block with the CX2CX6CX3C domain is marked by a black underscore, the Cys residues and the conserved Prolin are marked by a black spot.

Finally, each of the four leucines present in the LOB domain of LOB is marked with a underscore.

(23)

13 Byrne et al., 2000; Ori et al., 2000; Xu et al., 2003; Guo et al., 2008). Binding of AS2 to AS1 enables the binding of AS1 to the KNOX promoter region likely causing the formation of a repressive loop and chromatin remodeling complexes such as HIRA and Polycomb-repressive complex 2 (PRC2) are recruited (Phelps Durr et al., 2005; Guo et al., 2008; Lodha et al., 2013). PRC2 then induce a repressive chromatin state which is likely inherited through several cell divisions during leaf development as expression of AS1 and AS2 only overlap in young leaf primordia (Byrne et al., 2000; Xu et al., 2003; Phelps Durr et al., 2005; Guo et al., 2008; Lodha et al., 2013).

JLO are expressed in organ initiation sites and later in the boundaries between the Shoot Apical Meristem (SAM) and lateral organ primordia regulating the KNOX members SHOOT MERISTEMLESS (STM) important for SAM formation and maintenance and KNAT1 during lateral organ development (figure 1.5) (Borghi et al., 2007; Rast and Simon, 2012). Loss-of- function studies of JLO showed expanded expression of STM and KNAT1 across the base of lateral organ primordia in jlo indicating JLO as a negative regulator restricting KNAT1 and STM expression beyond SAM (Rast and Simon, 2012). In vivo analysis showed the ability of AS2 to bind to JLO, which could mediate the binding of JLO to AS1 as well (Rast and Simon, 2012). However, AS2 and JLO are also believed to act independently of AS1 as AS1 and AS2 show different phenotypical abnormalities when overexpressed and have overlapping but not identical expression pattern (Byrne et al., 2002; Iwakawa et al., 2002; Iwakawa et al., 2007;

Rast and Simon, 2012). Thus, JLO can relay its restrictive constrains on KNOX during plant development as part of a tetrameric complexes with AS2 and AS1 or a heteromeric complexes with AS2 or by forming a homomer (Rast and Simon, 2012).

Another in vivo study revealed that another LBD family member, LBD31, could interact with both AS2 and JLO (Rast and Simon, 2012). LBDs are obviously capable of forming complexes with each other as well as other transcription factors resulting in different binding abilities to different targets.

(24)

14

1.4.2 LBDs are in turn regulated by BOPs

In addition to regulate AZ formation BOP1 and BOP2 are important for leaf morphogenesis and patterning (Ha et al., 2003; Ha et al., 2007). BOP1 and BOP2 share distinct and overlapping expression and in vivo studies demonstrated their ability to dimerize (Norberg et al., 2005; Jun et al., 2010). BOP1 and BOP2 have been shown to be important for KNOX repression during leaf development as phenotypical defects of bop1 bop2 mutants are likely to be a result of ectopic KNOX1 expression (KNAT1, KNAT2 and KNAT6) as well as YABBY (YAB) expression, which is most likely important for maintaining the adaxial-abaxial polarity (Siegfried et al., 1999; Ha et al., 2003; Ha et al., 2007; Sarojam et al., 2010).

In vitro and in vivo studies show that BOP1 and BOP2 act as transcriptional activators of AS2 as demonstrated in figure 1.5 by direct binding of BOP1 to AS2 promoter region thus indirectly suppressing KNOX1 genes (Ha et al., 2010; Jun et al., 2010). Furthermore, genetic studies of stm embryos showed that ectopic expression of AS2 was dependant on active BOP (Jun et al., 2010). Other LBDs induced by BOP1 and BOP2 are LOB (figure 1.5), which can be directly activated by AS2-AS1 complex, and LBD36 both responding positive to increasing expression levels of BOP1 and BOP2 (Ha et al., 2007). However, BOP1 and BOP2 can also cause repression of KNOX genes in an AS2 independent manner as knox1 to some extent can rescue bop but not as2 phenotype (Ikezaki et al., 2010; Jun et al., 2010).

In inflorescence KNAT1 and PENNYWISE (PNY) are in turn believed to act as repressors of BOP1 and BOP2 as well as KNAT6 thus preventing premature cell differentiation and shortened internodes indicating that the timing of repression of KNOX is important for proper development (Ragni et al., 2008; Khan et al., 2012). Both inactivation of KNAT6 and BOP1/2 are able to partially rescue the defects of knat1 and pny placing KNAT6 and BOP1/2 in the same signaling pathway regulating inflorescence architecture and perhaps other pathways such as the IDA signaling pathway where KNAT1 also act as a repressor of KNAT6 inhibiting floral organ abscission (Ragni et al., 2008; Shi et al., 2011; Khan et al., 2012).

(25)

15 Figure 1.5: Simplified model of the genetic network underlying meristem patterning, organ-to-meristem boundary domain formation and organ initiation. Arrows and inhibition lines represent positive and negative regulations, respectively. LOB and the

tetrameric complex JLO/AS2/AS1 are active in the meristem-to-organ boundaries. Expression of LOB is regulated by BOP1/2 which also positively regulates AS2/AS1. JLO/AS2/AS1 acts as a suppressor of KNOX genes in the boundaries between meristem and organ primordia.

AS2/AS1 is active in the organ primordia where it interacts with chromatin remodeling

complexes to ensure repression of KNOX gene expression. Ha et al., (2010), Jun et al., (2010) and Žádníková (2014).

1.4.3 LBD37, LBD38, LBD39 and LBD41

LBD37, LBD38 and LBD39 arose most likely from two segmental duplications of the Arabidopsis genome (http://www.tigr.org/). LBD37, LBD38, LBD39 and LBD41, members of the Class II of the LBD family, are closely related LBDs belonging to the same cluster (Shuai et al., 2002). LBD38 and LBD39 are predicted to co-express in Arabidopsis thaliana and the LBD37 protein are found to be localized in the nucleus further supporting the role of LBDs as transcription factors (Rubin et al., 2009). LBD37, LBD38 and LBD39 have shown to be inducible by Nitrogen and thereby acting together as repressors of nitrogen responsive genes like PAP1/PAP2 causing a decrease in Anthocyanin synthesis (Rubin et al., 2009).

Interestingly, data provided by Meng et. al. (2010) show that LBD41 also localizes to the nucleus functioning as a transcription factor involved in determination on adaxial-abaxial identity in lateral organ development in the plant cockscomb (Celosia cristata) possibly

(26)

16

through regulations of KNOX members as malformed leaves and flowers seen in 35S:LBD41 probably were caused by misexpression of cockscomb KNOX genes (Meng et al., 2010).

In a yeast two-hybrid experiment LBD37 and LBD41 were found to interact with TOPLESS (TPL) and TOPPLESS-related (TPR) corepressors from the TPL/TPR family of Groucho/

Tup1-like corepressors (Causier et al., 2012). This opens a possibility of LBDs interacting with different cofactors as well as other TFs when relaying their regulatory constraints (Husbands et al., 2007; Causier et al., 2012).

As LBDs have been shown to be able to regulate KNOX genes in different plant developing processes and as LBD41 has been indicated a role as KNOX repressor in cocscomb it would be interesting to investigate the possibility of these four closely related LBDs acting as regulators of KNOX genes in floral organ abscission in Arabidopsis.

Figure 1.6: Phylogenetic tree over members of the LBD family class II. Genetic distance is included indicating the high degree of similarity between the class II LBD members.

1.5 Aim of study

In this study, we wanted to investigate the possibility of LBD37, LBD38, LBD39 and LBD41 acting in floral organ abscission. We propose here that the LBDs are negatively regulated upon the transduction of the IDA signal. This would inhibit them from suppressing class 1 KNOX Homeobox genes facilitating induction of floral organ abscission through activation of Cell Wall Remodeling (CWR) genes (figure 1.7). To test this proposed model and potential floral organ abscission related phenotypes single and double lbd mutant lines were made.

Different expression constructs were made in order to investigate the temporal and spatial

(27)

17 expression pattern of the LBDs. In addition, we wanted to examine the potential link between LBDs and factors of the IDA signaling pathway through genetic interaction studies by looking at the expression pattern of LBDs in ida and haehsl2. Furthermore, the potential rescue of the hae hsl2 phenotype by loss-of-LBDs was investigated. Finally, since negative regulation of KNOX genes by LBDs promotes differentiation of meristematic cells, the expression pattern and expression level of KNOX genes in lbd single mutant lines were analyzed.

Figure 1.7: Proposed model of IDA signaling pathway including LBD as inhibitor of abscission by repressing KNOX activity. The stippled lines symbolize regulatory

interactions that not yet have been confirmed. In this model IDA repress KNAT2 and KNAT6 through a proposed additional pathway where IDA positively regulates BOP1/2 which cause repression of LBDs, thus disabling repression of KNAT2 and KNAT6 leading to CWR gene expression. Modified from Shi et al., (2011).

(28)

18

2 Materials and Methods

2.1 RNA and DNA techniques

2.1.1 Isolation of genomic DNA

Genomic DNA, used to quantify promoter regions of LBD37, LBD38, LBD39 and LBD41 and as genomic control, was isolated following a protocol modified after Dellaporta et. al (1983).

20 milligrams (mg) of tissue from rosette leaves and sprouts were harvested from Colombia (Col) wild type (wt) and immediately frozen in liquid nitrogen ((l)N₂). The frozen tissues were crushed using a mortar. Preheated (65 °C) Elution buffer (EB) (100 mM Tris-HCl pH 8.0, 50 mM Ethylenediaminetetraacetic acid (EDTA) pH 8.0, 0.5 M NaCl, 1.25 % SDS, 8.3 mM NaOH, and 0.38 % Na bisulfate) containing mercaptoethanol (2-ME) (66 µl per 100 ml EB) was added. The solutions were transferred to corex tubes and incubated at 65 °C with occasional stirring. After 10 minutes (min) 5 ml 3 Molar (M) Potassium acetate (KAc) was firmly added and the solutions were left on ice for 20 min followed by a 10 min centrifugation at 4 °C, 10 000 revolutions per min (rpm), using a TJ-25 centrifuge (Beckman Coulter, Inc.) with a TS-5.1-500 rotor. Supernatants were filtered through mira cloth into a new corex tube with isopropanol and left to precipitate for 1 hour (h) at room temperature (RT) followed by a 20 min centrifugation at 4 °C, 10 000 rpm. The pellets were cleansed in 70 % Ethanol (EtOH) twice before resolving it in T5E (50 mM Tris-HCl pH 8.0, 10 mM EDTA) over night (O.N) at 4 °C. The day after 14 µl RNase were added to the resolved pellets and incubated at 37 °C for 1 h. 50 µl 3 M Natrium acetate (NaAc) per 500 µl of resolved pellets together with pre cooled (-20 °C) absolute alcohol (1400 µl pr. 500 µl) were added. After centrifugation supernatants were removed and pellets were cleansed twice with 70 % EtOH. When dry, the pellets were resolved in 250 µl distilled water (dH2O) O.N at 4 °C.

(29)

19

2.1.2 DNA isolation with Ultraprep Genomic DNA plant Kit

For isolation of genomic DNA for genotyping Ultraprep Genomic DNA Plant Kit from AHN Biotechnologie GmbH was used and the protocol from the manufacturer was followed. 150 mg of Arabidopsis leaves were frozen in (l) N2 before homogenization using Retch MM301(Retch GmbH). PB buffer with Proteinase K were added to the homogenized tissue and the mixture was vortexed before incubation at 52 °C for 30 min followed by a 5 min centrifugation at RT, 13 000 rpm. Clarified supernatant was transferred to a new Eppendorf tube and AB was added. The solutions were transferred to a spin column and the samples were washed with Wash Buffer (WB) and 70 % EtOH before the DNA was eluted with Elution buffer (EB) buffer.

2.1.3 RNA isolation

Isolation of RNA was done using Spectrum™ Plant Total RNA Kit (Sigma-Aldrich). As the mRNAs are quickly degraded the tissue was stored at -80 °C. Siliques at position 4, 6 and 8 along the inflorescence (position 1 representing the youngest flower with visible white petals (Bleecker and Patterson, 1997), from 10 plants were collected in Eppendorf tubes and frozen in (l) N2. The frozen tissue was transferred to an Eppendorf tube containing small ceramic beads and 500 µl lysis solution (10 µl 2-mercaptoethanol (2-ME) for every 1mL lysis buffer) was added on ice. The tissue was crushed using a MagNAlyser (Roche) at 7000 rpm for 15 seconds (sec), followed by a cool-down at -20 °C for 2 min and finally the samples were centrifuged for 1 min at 4 °C, 13 000 rpm using a Eppendorf 5415R Refrigerated Micro Centrifuge. These steps were repeated until the tissue was completely homogenized. The sample was transferred to a sterile Eppendorf tube and centrifuged for 3 min at 4 °C, 13 000 rpm before supernatant was transferred to a Filtration column and centrifuged for 1 min at 4

°C, 13 000 rpm. Binding solution was added to the clarified flow- through and mixed by pipetting carefully before transferring the flow-through onto a Binding column. To remove any DNA On-Column DNase digestion was performed using On-Column DNase Kit (Sigma- Aldrich). 80µl of digestion solution (10 µl DNase in 70 µl DNase buffer) was added to the center of the Binding column. To remove the digested DNA the Binding column was washed with Wash solution 1 and for a final wash Wash Solution 2 diluted with EtOH was added.

(30)

20

When all liquid was removed, the column was transferred to a sterile 2 ml collection tube.

Purified RNA was eluted with 50 µl nuclease-free water and stored at – 20 °C.

2.1.4 cDNA synthesis

After isolating the RNA it was transcribed into complementary DNA (cDNA) using SuperScript® III Reverse Transcriptase (Invitrogen). Oligo dT primers contain a poly T-tail which bind to the poly A tail in the 3’end of the mRNA. By addition of dNTP, reverse transcriptase (RT) buffer and RNase inhibitor together with RT the enzyme will reversely transcribe the mRNA into double stranded cDNA. To remove template RNA RNase H was added at the end of the inactivation of the RT enzyme.

2.1.5 Polymerase chain reaction

Polymerase chain reaction (PCR) is a fast and easy method for detecting and amplifying DNA or RNA fragments in vitro. Key components in a PCR reaction are a thermostable DNA polymerase for synthesizing the new DNA from the template, dNTPs (deoxynucleotide triphosphate), primers which provides a free 3’ hydroxyl group onto which the DNA polymerase can attach new dNTP. The PCR program has several steps: denaturation of the template into single stranded DNA (ssDNA), primers annealing to the ssDNA and elongation leading to synthesis of complementary strands. Variations of a general PCR program (table 2.1) were used, varying annealing temperature depending on the primers and the annealing time, amplification temperature and amplification time depending on the polymerase used.

When working with primer sets with different annealing temperature touchdown programs were used, lowering the temperature 0.2 – 0.3 °C every cycle.

The promoter region of LBD37, LBD38, LBD39 and LBD41for cloning (section 2.1.13) were amplified using proofreading polymerases, Advantage 2 polymerase (Clontech) and KOD Hot Start DNA polymerase (TOYOBO). Screening for correct insert of purified plasmid DNA (section 2.1.14) from transformed bacteria (section 2.2.2 and 2.2.3) and for Real-Time PCR of the single mutant lines Taq DNA polymerase (New England BioLabs) was used, as

(31)

21 proofreading abilities were not required. For genotyping of plant lines (section 2.2.12 and 2.1.2) RedExtract readyMix (Sigma -Aldrich) or Taq DNA polymerase was used. The PCR setup was done according to the manufacturers recommendations. Primers used are listed in appendix 2.

Table 2.1. Standard PCR program. The period for the amplification step depends on the rate of the polymerase and the length of the wanted product. Annealing temperature depends on the optimal binding temperature for the primers used. When doing PCR on bacterial cultures the first denature step is set to last for at least 8 min to destroy the cell membrane so the DNA is released.

Process, temperature (°C) Time Cycles

Denaturation, 95 °C 8-3 min 1

Denaturation, 95 °C 30 sec

30-40 cycles Annealing 55-70 °C 30 sec

Amplification, 68-72 °C 1-3 min

Final extension, 68-72 °C 5-10 min 1

Hold, 4-10 °C ∞ -

2.1.6 PCR clean-up system

For cleaning of PCR products and cDNA the Wizard^® SV Gel and PCR Clean-Up System (Promega) was used following the protocol provided by the manufacturer. Equal amounts of Membrane binding solution as PCR amplifications were mixed and applied onto a SV Minicolumn assembly. After centrifugation for 1 min at RT, 13 000 rpm, the column with the DNA was washed using Membrane solution twice. To remove the last traces of EtOH the samles were placed at 37 °C for 10 min. PCR products and cDNA were eluted with 50 µl and 40 µl nuclease-free water, respectively.

(32)

22

2.1.7 Quantitative RT-PCR

Real time quantitative PCR (RT-qPCR) allows amplification and detection of the PCR product simultaneously, and is used in gene expression analysis, miRNA analysis, genetic variation analysis and protein analysis. The qPCR machine has the amplification factors of a regular PCR machine as well as the capability to detect fluorescence light.

Expression of KNAT1, KNAT2 and KNAT6 in lbd single mutants and LBD37 and LBD39 in ida and hae hsl2 mutants were measured using the LightCycler^® 480 (Roche) and LightCycler^®96 (Roche) following the setup-recommendations provided with the SYBR®

Green I Master dye Kit (Roche).

The SYBR® Green I Master dye send out fluorescence light when bound to double-stranded DNA. Under denaturation of double stranded DNA the SYBR^® Green is released causing drastic reduction in the fluorescence. During the polymerization, primers anneal starting the generation of newly synthesized PCR product. After the polymerization, more SYBR^® Green I Master dye will be able to bind due to the amplification increasing the detectable fluorescence. Due to the molar excess of primers and thermostable DNA polymerase at the beginning, the DNA template will be the limiting substrate in the reaction making the fluorescence detection proportional to the amount of DNA template. After a certain amount of cycles, a threshold level is reached giving rise to a Cp value (crossing point/ crossing threshold) reflecting the amount of DNA template present in the samples. The earlier the threshold level is reached the lower will the Cp value be, indicating a higher amount of template to begin with. Primers used are listed in appendix 2.

2.1.8 Reverse Transcriptase-PCR

After converting mRNA to cDNA the quantity of RNA can be detected by PCR using suitable primers which can amplify the cDNA of interest. The amplified cDNA is then run on an agarose gel where the size and amounts of cDNA are estimated.

(33)

23

2.1.9 Gel electrophoresis

PCR samples were checked by agarose gel electrophoresis, which separates the DNA or RNA fragments according to size. Depending on the expected sizes of the fragments different percentages of a SeaKem^® Agarose was used in 1 X TAE (Tris acetate, EDTA). Ethidium Bromide (EtBr) was added to the gel and the results were visualized under UV-light. The more fragments of a given size the stronger the band will appear when visualized giving an indication of the amounts of a given fragment. The gels were run on 85-100V for 30-60 minutes depending on the size of the gel. 6 X FBX loading dye (15 % Ficoll 400 (GE Helthcare), 0.25 % Orange G (Sigma), dH2O) was added to the samples to a final concentration of 1 X FBX before loading the samples on the gel. For determination of the size the DNA standard 1Kb ladder (Thermo Scientific) was loaded on the gel.

2.1.10 Quantification of DNA and RNA

Quantification of DNA and RNA was done using the NanoDrop^® 2000 UV-Vis Spectrophotometer (Thermo Scientific) or NanoDrop^® ND-1000 UV-Vis Spectrophotometer (Thermo Scientific). The NanoDrop instrument calculates the DNA or RNA concentration based on the Beer-Lambert equation: A = ɛ *b * c where A is the absorbance (RNA and DNA have A260), ɛ is the extinction coefficient (L * mol^-1* cm^-1), b is the length the UV travels in cm, and c is the concentration (moles/liter).

2.1.11 Sequencing

Sequencing was performed by the ABI laboratory at the Department of Bioscience, Faculty of Mathematics and Natural Sciences, University of Oslo N-0316.

(34)

24

2.1.12 Genotyping of SALK-lines

For genotyping mutant lines the REDExtract-N-Amp Plant PCR Kit (Promega-Aldrich). This kit allows rapid DNA extraction and PCR preparations. Extraction buffer was added to leaf discs from plants to be analyzed and incubated at 95 °C for 10 min before Dilution buffer was added. Extracted DNA was stored at 4 °C. PCR on extracted DNA was run using REDExtract-N-Amp™ PCR ReadyMix™ (Promega-Aldrich) and genotyping primers (appendix 2).

2.1.13 Gateway

^®

cloning system by Invitrogen

The Gateway^® cloning system is based on the site-specific recombination system of bacteriofag lambda ( λ) enabling it to transfer DNA to the Escherichia coli genome (Landy, 1989). The site-specific attachment sites, att sites, allow insertion of a DNA sequence into a vector without altering the orientation and reading frame as the attB1 site only will recombine with attP1 and not attP2 and attL1 only will recombine with attR1. The recombination of these sites is mediated by the BP clonase™ enzyme which recombines attB sites the DNA sequence with attP sites in the vector resulting in entry clones with attL sites. An LR clonase™ enzyme recombine the attL sites with attR sites in the destination vector of your choice.

By using primers with flanking attB sites the promoter region of LBD37, LBD38, LBD39 and LBD41 with flanking attB sites were quantified. (Primer sequences used for quantification of LBD promoter regions with flanking attB sites are listed in appendix 2). The attB flanked PCR products were then recombined with the attP sites of the donor vector pDONR™/Zeo (Invitrogen) resulting in an entry clone and a by-product. The attL sites of the entry clone containing the promoter region were then recombined with attR sites in the destination vectors pMDC162 (section 2.1.15) and pHGY (RIKEN Plant Science Center) for the making of expression clones with the reporter gene β-Glucoronidase (GUS) and Fluorescent Protein (YFP), respectively (figure 2.1). The donor vector and the destination vectors had a ccdB gene as a negative selection marker that ensured the survival of E. coli cells containing the promoter region and not the ccdB gene, as the ccdB interferes with the DNA gyrase activity killing the E. coli cells. As a positive selection marker, the donor vector encoded a zeocin (zeo) resistance gene, while the destination vector pMDC162 encoded a kanamycin (Km)

(35)

25 resistance gene and pHGY encoded a Spectinomycin (Sp) resistance gene. The antibiotic resistance ensured the survival of the E. coli cells containing the vector of interest on plates with the respective antibiotic. The BP reaction and LR reaction were performed according to the manufacturer’s recommendations. For termination of the BP and LR reaction, 1 µl Proteinase K (Invitrogen) was added as the enzyme degrades the reaction enzyme. Solutions were incubated at 37 °C for 10 min.

Correct insertion of the respective promoter regions was confirmed by sequencing using the same primers that was used for amplification of the LBD promoter regions (for primer sequences see appendix 2).

Figure 2.1: Illustration of the two LR vectors with the integration of the LBD promoter region upstream of the reference gene. Both vectors supply the infected plants with hygromycin resistance (green arrow) enabling easy selection of transformants.

(36)

26

2.1.14 Plasmid DNA purification of lysate

Plasmid DNA from E. coli was purified using the Wizard®Plus SV Minipreps DNA Purification System Kit (Promega) following the centrifugation protocol provided with the kit. After making bacterial lysate the cleared lysate was transferred to a column inserted into a collection tube followed by centrifuged. After discarding of flow-through the membrane in the column was washed twice with Wash solution previously diluted with EtOH. Plasmid DNA was eluted into a new sterile Eppendorf tube with nuclease-free water.

2.1.15 Midiprep Plasmid purification

For recovering a larger amount of plasmid DNA, a JetStar™ 2.0 Plasmid Kit-Midi (Genomed of Life Technologies Corporation) was used and the recommended protocol from the manufacturer was followed. The Midiprep was performed on bacteria pellet of bacteria with pMDC162 vector containing a Km resistance gene. The pellet was resuspended in resuspension solution, lysed with lysis solution followed by neutralization of the lysate by the addition of Neutralization solution. Supernatant was run through a previously equilibrated column by gravity flow. Column was the washed with Column wash solution. Plasmid DNA was eluted by addition of preheated DNA elution solution. Isopropanol was added to precipitate the plasmid and the pellet containing plasmid DNA was washed with 70 % EtOH.

When totally free of EtOH the pellet was resuspended in MilliQ-water and stored at -20 °C.

(37)

27

2.2 Bacterial techniques

2.2.1 Bacterial growth conditions and selection

2.2.1.1 Growth and selection of E.coli

Optimal growth temperature for E. coli is 37 °C and E. coli has a generation time of about 25 min depending on the E. coli strain. Inoculation time was between 16-20 h before reaching the stationary phase. For selection, different antibiotics were used depending on the antibiotic resistance of the transformed E. coli. Transformed E. coli was always grown in liquid or on solid LB (10 g/l Peptone, 5 g/l Yeast extract, 10 g/l NaCl) media containing the respective antibiotics.

2.2.1.2 Growth and selection of A. tumefaciens

The optimal growth temperature for A. tumefaciens is 28 °C and has a generation time of about 1 h depending on the strain. Inoculation time was around 48 h before reaching stationary phase. For selection antibiotics, to which the Agrobacteria were resistant, as well as the antibiotic resistance provided by the plasmid DNA, was added to both the solid growth media and the liquid YEB media. (YEB = 5 g/l Beef extract, 1 g/l Yeast extract, 1 g/l Bacto Peptone, 5 g/l sucrose, pH set to 7,4 and after autoclavation addition of 2 ml 1 M MgSO4).

2.2.1.3 Freezing stock

Freezing stocks were made for all of the different transformed bacteria cultures: 50/50 of bacterial culture and 80 % Glucose solution was mixed before storing at - 80 °C.

(38)

28

2.2.2 Escherichia coli transformation by heat shock

E. coli transformation was performed by heat shock. 1 μl plasmid DNA was added to a one shot DH5α™ Chemically Competent cells (Invitrogen) or One Shot® Top10 Chemically competent cells (Invitrogen) and heat shocked at 42 °C for 40 sec, just enough to make the cell wall permeable for the surrounding plasmid DNA without killing the cell. After addition of SOC medium, the transformed E. coli cells were shaken at the optimal growth temperature, 37 °C, for 1 h before plating out on plates containing LB medium with the respective antibiotics. For E.coli with pDONRzeo vector zeo was added to a final concentration of 2.5 ng/μl. For the GUS constructs, using pMDC162 with Km resistance, Km were added to the growth media to a concentration of 100 ng/μl, while for the YFP constructs using pHGY vectors with Sp resistant, Sp were added to a concentration of 50 ng/μl.

2.2.3 Agrobacterium tumefaciens transformation by electroporation

For transformation of A. tumefaciens electroporation was used. Here the plasmid DNA was added to a one shot of electro competent C58 GV3101 or C58 GV2260 Agrobacteria cells.

An electrical shock was applied to the reaction creating pores on the cell wall of the agrobacteria making it possible for the surrounding plasmid DNA to enter the cells. SOC medium was added after the electroshock and the cells were left shaking at 28 °C for 1-2 h to recover before plating out on previously made YEB plates containing the respective antibiotics; Km for GUS constructs and Sp for YFP constructs. In addition to the antibiotic resistance provided by the different constructs GV3101 Agrobacteria strain had Rifampicin (Rif), Gentamycin (Gent) and Km resistance while the GV2260 strain had Rif and Gent resistance. GV3101 was used for the YFP constructs while GV2260 was used for the GUS constructs. Km and Rif were added to a final concentration of 100 ng/μl, while Gent was added to a final concentration of 7 ng/µl.

(39)

29

2.2.4 Preparation of Agrobacteria solution for floral dipping

For floral dipping the transformed agrobacteria were cultured in a 500 ml Erlenmayer flask containing YEB medium and the respective antibiotics in a total volume of 250 ml. The cultures were inoculated shaking at 28 °C for about 48 h depending on how fast the agrobacteria was growing before measuring the OD. OD had to be no more than 1.2 before proceeding with the making of the infiltration culture keeping the bacteria in the stationary phase. If the OD₆₀₀ ≥1.2 the bacterial culture was diluted to an OD lower than 1.2 and incubated until the OD reached 1.2 again. When OD ≤ 1.2 the cultures were centrifuged using a TJ-25 centrifuge (Beckman Coulter, Inc.) with a TS-5.1-500 rotor for 10 min at RT, 5000 rpm. Pellets were then dissolved in 5 % Sucrose solution until it reached OD₆₀₀= 0.8. Before the floral dipping, Silwet L-77 (Lehle Seeds) was added to a total of 0.02 %. This was to make the infiltration easier for the Agrobacteria.

2.2.5 Production of cleared bacteria lysate

For lysing the bacterial culture, the protocol for the Wizard^®Plus SV Minipreps DNA Purification System Kit (Promega) was followed. The bacterial cultures were pelletized by centrifugation using a table centrifuge. The pellet was resuspended by using Cell Resuspention Solution. The bacteria cells were lysed by the addition of Cell Lysis Solution.

Alkaline Protease solution was added to digest the proteins in the bacterial cell before the addition of neutralization solution. Cleared lysate for further purification of plasmid DNA was gained by centrifugation to separate the precipitated digested proteins from the plasmid DNA in the supernatant.

(40)

30

2.3 Plant techniques

2.3.1 Seed sterilization and growth conditions

Before plating out seeds they were sterilized to get rid of mold and bacteria that may follow the seeds. 70 % Ethanol was applied to the tubes containing the seeds and incubated at room temperature for 5 min. EtOH was poured off and Bleach solution containing 20 % Klorix in 0,1% Tween 20 and double-distilled H2O (ddH2O) was added. After 5 min, Bleach solution was poured off followed by addition of wash solution. Finally 0,1 % agar were added before pouring the seeds onto the MS-2 (Murashige & Skoog Medium, sucrose, KOH to set the pH to 6,3 and Bacto Agar) plates with or without antibiotics depending on whether the seeds were expected to have a resistance or not. Seeds were distributed on the plates by rotation. Plates were sealed off using surgical tape. The plants were placed in the growth room at 22°C under Long Day conditions with light intensity 100 µE/m2 after 1-2 days at 4 °C in the dark. After 10-14 days the seedlings were transferred to soil keeping the plants at 18 °C under Long Day conditions. For harvesting seeds older plants were placed in a harvesting room to dry before collecting the seeds.

2.3.2 Transformation of Arabidopsis thaliana

Col-0 plants for transformation were trimmed before transformation to get more branches for the floral dipping. When the plants had recovered from the cutting and gained more branches the plants were held upside down and dipped into the solution containing transformed Agrobacterium tumefaciens and Silwet L-77. After infecting the plants with the transformed Agrobacteria, the plants were laid down on wet paper towels in a tray and wrapped in aluminum foil O.N in the dark before transferring them back in the growth room for recovery and further growth.

(41)

31

2.3.3 Selection of GUS and YFP lines

After transformation seeds (T1) from the transfected plants (T0) were collected and plated out on plates containing the respective antibiotics (hyg in both cases). Plates with a 3:1 segregation of transformed: non-transformed plants was selected as these are hemizygous. To find homozygous lines in the T₃generation seeds from the different hemizygous T₂lines were plated out to find the plates with only transformed plants indicating homozygous lines.

2.3.4 Histochemical GUS analysis

Histochemical GUS analyses were performed on proLBD::GUS lines in wt, ida and haehsl2 as well as proKNAT::GUS lines in lbd38lbd39. Seedlings and the floral positions 2, 4, 6, 8, 10, 12, 14 and 16 from the T1 and T2 generation were stained with the substrate 5-bromo-4- chloro-3-indolyl ß-D-glucuronide (X-Gluc). At 37 °C X-Gluc is cleaved by β-Glucoronidase into glucoronic acid and chloro-blomoindigo. When oxidized the chloro-blomoindigo dimerize giving a blue colored precipitate that is visible in regular light. Although the optimal temperature for GUS A is 37 °C the proLBD::GUS lines were incubated at RT due to the highly active promoters.

Before staining the different tissues were harvested into wells containing ice cold 90 % Acetone. The staining was done by incubation of the tissue in the staining buffer (5ml 500mM NaPO4 pH7.2, 1 ml 100mM K4Fe(CN)6 x 3H2O, 1ml K3Fe(CN)6, 0,5 ml 10% Triton X-100, 42,5 ml dH2O) containing 1ml 100mM X-gluc in dimethylformamide (DMF) for 0,5-20 h.

After staining the tissues were dehydrated by incubation in a graded EtOH series to 70 % (15

% EtOH, 35 % EtOH and 50 % EtOH solutions were diluted with 50mM NaPO4) before fixation with FAA solution (10 ml EtOH, 7 ml dH2O, 2 ml Glacial Acetic Acid and 1 ml 37 % Formaldehyde). After fixation the tissues were rehydrated by incubating the tissues in a reversed graded EtOH series to 0 % EtOH.

The samples were mounted on microscopy slides covered with a few drops of clearing solution (8g Chloral hydrate, 2 ml dH2O and 1 ml 87 % Glycerol). The specimen was left to clear at 4 °C for at least 1 h before inspection.

The role of LBD proteins in floral organ abscission